This article was downloaded by: [Francesca Alby]On: 23 January 2015, At: 11:05Publisher: RoutledgeInforma Ltd Registered in England and Wales Registered Number: 1072954 Registeredoffice: Mortimer House, 37-41 Mortimer Street, London W1T 3JH, UK

Click for updates

Mind, Culture, and ActivityPublication details, including instructions for authors andsubscription information:http://www.tandfonline.com/loi/hmca20

Diagnostic Decision Making in Oncology:Creating Shared Knowledge andManaging ComplexityFrancesca Albya, Cristina Zucchermaglioa & Mattia Baruzzoa

a Sapienza University of RomePublished online: 21 Jan 2015.

To cite this article: Francesca Alby, Cristina Zucchermaglio & Mattia Baruzzo (2015): DiagnosticDecision Making in Oncology: Creating Shared Knowledge and Managing Complexity, Mind, Culture,and Activity, DOI: 10.1080/10749039.2014.981642

To link to this article: http://dx.doi.org/10.1080/10749039.2014.981642

PLEASE SCROLL DOWN FOR ARTICLE

Taylor & Francis makes every effort to ensure the accuracy of all the information (the“Content”) contained in the publications on our platform. However, Taylor & Francis,our agents, and our licensors make no representations or warranties whatsoever as tothe accuracy, completeness, or suitability for any purpose of the Content. Any opinionsand views expressed in this publication are the opinions and views of the authors,and are not the views of or endorsed by Taylor & Francis. The accuracy of the Contentshould not be relied upon and should be independently verified with primary sourcesof information. Taylor and Francis shall not be liable for any losses, actions, claims,proceedings, demands, costs, expenses, damages, and other liabilities whatsoever orhowsoever caused arising directly or indirectly in connection with, in relation to or arisingout of the use of the Content.

This article may be used for research, teaching, and private study purposes. Anysubstantial or systematic reproduction, redistribution, reselling, loan, sub-licensing,systematic supply, or distribution in any form to anyone is expressly forbidden. Terms &

Conditions of access and use can be found at http://www.tandfonline.com/page/terms-and-conditions

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

Mind, Culture, and Activity, 00: 1–19, 2015Copyright © Regents of the University of California

on behalf of the Laboratory of Comparative Human CognitionISSN 1074-9039 print / 1532-7884 onlineDOI: 10.1080/10749039.2014.981642

Diagnostic Decision Making in Oncology: Creating SharedKnowledge and Managing Complexity

Francesca Alby , Cristina Zucchermaglio, and Mattia BaruzzoSapienza University of Rome

Drawing upon the concept of practice, the article explores diagnostic decision making in oncologythrough the analysis of informal conversations between doctors in an Italian hospital. The analysisshows that doctors rely on three collaborative practices: (a) joint interpretation, (b) intersubjectivegeneration and validation of hypotheses, and (c) postponing the diagnostic decision. Through suchpractices, doctors jointly handle tough issues such as managing complexity, dealing with cognitivedifficulties and limits of knowledge, and avoiding diagnostic errors. The article addresses some lacu-nae in the literature by providing empirical access to how decision making is done in an understudiedand specialized branch of medicine.

INTRODUCTION

The article analyzes diagnostic decision making in oncology. Owing to its complexity, oncologyis a particularly interesting context for analyzing medical thinking and doing. It is a branch ofmedicine that deals with a systemic illness with a complex pathogenesis: an illness that can extendto every organ of the body with a fatal outcome. Although it is a discipline with a high rate ofdevelopment and scientific innovation, it is also a field—as a system of formalized and validatedknowledge—in which there are wide margins of uncertainty (Han, Klein, & Arora, 2011). Howdo doctors manage this complexity and epistemic uncertainty? How do they make decisions in ill-structured situations? How do they keep the risk of diagnostic errors under control? In this article,we address these questions through analysis of a corpus of audio-recorded informal conversationsbetween doctors collected in an Italian hospital. This study is, to our knowledge, the first to useempirical data of this kind to analyze these topics; thus, we believe it can furnish novel insightsinto diagnostic decision making in specialized medical domains.

In the study of medical diagnosis, decision making models and epistemologies have tradi-tionally been informed by cognitivist and individualistic perspectives (see Han et al., 2011;

Correspondence should be sent to Francesca Alby, Sapienza University of Rome, Department of Developmental andSocial Psychology, via dei Marsi, 78, Rome 00185, Italy. E-mail: francesca.alby@uniroma1.it

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

2 ALBY, ZUCCHERMAGLIO, AND BARUZZO

Kostopoulou, Russo, Keenan, Delaney, & Douiri, 2012) that represent such processes as rationaland linear:

Doctors (and patients) are constantly called on to make decisions. To do so, they must iden-tify the relevant information (such as the outcome of a clinical examination), develop a judgment(e.g., a diagnosis), choose a course of action among several possible on the basis of their preferences(e.g., a particular surgery), and then take action. (Motterlini & Crupi, 2005, p. 13; English translationby the authors)

On the basis of this model, medical diagnosis and decision making are then considered tobe abstract mental processes that take place “before” practical action and independent of theprofessional domain in which these processes of choice occur.

This approach is consistent with rational choice theories that are widespread references inthe interpretation of the decision making processes in many fields. The impact that formal logichad on research and theory on reasoning processes is also another factor that contributed to pushdecision making toward an idealized notion of pure rationality (Falmagne, 2000). To date, severalauthors have challenged such a model of linear rationality, highlighting a number of criticismsand limiting assumptions. As Oliver and Roos (2005) pointed out, most cognitivist studies intodecision making rely on some key notions. The first is the notion that decisions are moments intime in which a course of action is selected. The second assumption is that decisions are rational,or logically consistent, purely cognitivist “if–then” decision processes. Third, decision makingis frequently assumed to be a context-independent activity. With some exceptions, most of thisresearch has involved the study of individuals in laboratory, rather than naturalistic settings (Fiske& Taylor, 1991).

Simon (1957) was among the first who showed that people seek solutions they considergood enough in a given situation, making decisions that are based on bounded rationality. Morerecently, researchers studying organizational decision making and naturalistic decision makingsituated decision makers within organizational contexts, exploring the structures and social normsthat shape expectations and behaviors (Lipshitz, Klein, & Carroll, 2006; March, 1994; Szambok& Klein, 1995).

In line with these studies of decision-makers-in-context, the perspective adopted in this articleviews diagnostic decision making as a fundamentally social activity constituted in interactionand situated within real-world settings (Engeström & Sannino, 2012; Lave, 1988; Resnick, Säljö,Pontecorvo, & Burge, 1997; Scribner, 1986; Vygotsky, 1990). This perspective has a key defin-ing characteristic in its reference to the concept of practice. Whereas practice-based approachesshare family resemblances, “there is no unified practice approach” (Schatzki, 2001a, p. 11), noris there agreement on what counts as a practice (Nicolini, 2012). Bernstein (as cited in Miettinen,Paavola, & Pohjola, 2012) proposed a distinction between “classical” and “new” approaches topractice related to different philosophical traditions. In this categorization, classical practice the-ories draw upon the dialectical tradition of Marx and Hegel, whereas new approaches draw uponthe work of Heidegger and Wittgenstein. Within the classical tradition, cultural-historical activitytheory (Engeström, Miettinen, & Punamäki, 1999) and a sociocultural perspective on practice(Chaiklin & Lave, 1993) have focused, among other things, on work conceived as an impor-tant human activity in which people “simultaneously create both themselves and their materialculture” (Miettinen et al., 2012, p. 346).

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

DIAGNOSTIC DECISION MAKING IN ONCOLOGY 3

According to this approach, cognition and action are both shaped by participation in purpose-oriented and socially organized systems of activity mediated by the tools available. Mediation isa key concept in the “classical” approach to practice that has its roots in the work of the Russianpsychologist Lev Vygotsky. In Vygotsky’s thought, mediation by tools and signs as culturalartifacts—first experienced by participating in social activity with others—is gradually inter-nalized, thus becoming “internal speech” and individual thought. Internal cognitive processesare, therefore, social in origin, and they continue to operate through cultural tools for thinking.In workplaces, tools such as “scientific concepts” (as those of medicine), evidence from the dis-ciplinary literature, and specialized language are used to reason and to make decisions in routinework activities (Hutchins, 1993; Lave, 1993).

We position our analysis of diagnostic decision making within the previously outlinedsociocultural perspective on practice that considers cognition to be situated and distributed.We consider as “practice” what Schatzki (2001b) defined as “a set of actions,” for example,“farming practices comprise such actions as building fences, harvesting grain, herding sheep,judging weather, and paying for supplies” (p. 56). In what follows, we refer to “practices”as the set of actions through which doctors make diagnostic decisions in a system of special-ized medical activity such as oncology. Moreover, we combine in the term “practice” both thecultural/contextual dimension of actions and the cognitive one. On one hand, we consider this setof actions to be situated, mediated, and dependent upon the constraints and resources of the localdomain of activity (i.e., oncology). On the other hand, we consider such actions as loci in whichto study cognition as a social and cultural phenomenon (Cole, 1996; Goodwin, 1994; Hutchins,1995; Lave, 1988). As Sylvia Scribner (1984) well pointed out, “The general construct of prac-tice offers a possibility for integrating social-cultural and psychological levels of analysis andachieving explanatory accounts of how basic mental processes and structures become specializedand diversified through experience” (p. 13). In our approach, practice is, therefore, the key todescribing the complexity of human cognition and its connections with the everyday activities inwhich it occurs (see also Resnick et al., 1997).

THE RELEVANCE OF DIAGNOSTIC COMMUNITIES

This view of practice is consistent with Lave and Wenger’s (1991) construct of “community ofpractice.” This construct well represents the importance of social life in shaping local culturesand ways of doing and thinking. In Wenger’s (2002) definition:

A community of practice is a group of people who share an interest in a domain of human endeavorand engage in a process of collective learning that creates bonds between them: a tribe, a garage band,a group of engineers working on similar problems. (p. 2341)

With this construct, Lave and Wenger (1991) located knowing, reasoning, decision making, andother cognitive phenomena in the practices shared by a community of practitioners.

Empirical studies have described the importance of communities of practice in man-aging diagnostic work (see Gherardi, 2012). Viewed through a “practice lens,” diagnosticwork has proved to be shaped by organizational goals and tacitly shared priorities (Alby &Zucchermaglio, 2009; Boden, 1994; Button & Sharrock, 1998; March, 1991). Emblematicis Orr’s (1996) study, which exemplified how the diagnostic process for expert technicians

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

4 ALBY, ZUCCHERMAGLIO, AND BARUZZO

was essentially narrative and cooperative. Orr showed that the interpretation of technology-mediated events is an inevitably problematic and discursive activity and that it constitutesa crucial component of a community’s professional competence. It is, in fact, through jointaccounts and interpretations that technicians construct a repertoire of distributed knowledge andpragmatic understanding, which is one of the most valuable and enduring outcomes of theircollaboration.

Similarly, Toulmin (1996) highlighted the importance of doctors’ experience for diagnosticreasoning and decision making: “The everyday skills of experienced doctors today take thembeyond the limits of theory” (p. 211). In particular, he argued that there are not “typical” casesto be diagnosed, but situations that resemble one another only partially. Toulmin stated that doc-tors then need to consider the family resemblances among the varied patients’ conditions, whichextends beyond what he called a “theory of illness”: “Patients go to doctors with many variedconditions, for pragmatic not theoretical reasons: the doctor first sorts them into those in urgentneed of surgery, those who simply need a rest and a good meal, and so on” (p. 214). In so doing,doctors follow a situated rationality that enables them to deal with the complexity and localspecificities of their diagnostic work.

Other naturalistic studies have highlighted that there are features of processes such as diag-nosis, problem solving, and decision making that are not visible at an individual level andthat emerge only with a detailed analysis of the community’s work practices (see Alby &Zucchermaglio, 2006; Goodwin & Goodwin, 1996; Hutchins & Klausen, 1996; Roth, Multer,& Raslear, 2006). In particular, talk proves to be a key resource for dealing with problematicaspects of ongoing activities and for the organization of both aspects of distributed cognition andsocial organization of work (see Alby & Zucchermaglio, 2008; Boden, 1994; Goodwin, 1994;Hutchins, 1995; Streeck, Goodwin, & LeBaron, 2011; Zucchermaglio & Alby, 2012).

During some of the first ethnographies of medical activities, Cicourel (1985, 2002) observedthe distinctive character and function of talking in organizing diagnostic reasoning and clini-cal understanding in medicine. He showed the discursive strategies with which more or lessexperienced doctors combine formalized medical knowledge with tacit, procedural knowledgeand adjust them to the patient’s case. Based on conversation analysis and ethnomethodolog-ical perspectives, an extensive literature has developed focusing on medical interactions (fora collection of studies, see Beach, 2012; Heritage & Maynard, 2006; Pilnick, Hindmarsh, &Gill, 2009). These studies have dealt with, among other things, the study of diagnosis and deci-sion making in various medical fields (see Antaki, Barnes, & Leudar, 2005; Brookes-Howell,2006; Heath, 1992; Koenig, 2011; Maynard, 1992; Opel et al., 2013; Peräkylä, 1998; Stivers,2002).

The practice-based approach adopted here is different from this conversation analysis litera-ture in two main respects: (a) the analyses intend to be more grounded on the many constraintsand affordances of modern medicine as a set of specialized and complex institutional prac-tices, and (b) the analyses take account of hospital diagnostic communities, going beyond thedyadic relationship between doctor and patient during medical examinations. Our study focuseson physicians in a hospital who share a history of diagnostic interactions as the basis, and the out-come, of a practice of talking to each other and sharing knowledge for diagnostic purposes. In thisarticle, we explore in particular the practices through which doctors organize their diagnosticdecision making during joint conversations about cancer cases.

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

DIAGNOSTIC DECISION MAKING IN ONCOLOGY 5

RESEARCH METHOD, DATA CORPUS, AND ANALYSIS PROCEDURES

The research was conducted in a medium-sized public hospital serving a major city in centralItaly. It had more than 150 overnight beds and more than 15 day-hospital beds. It employedabout 800 workers. Cancer care was one of the many services delivered at the hospital.1 Wecollected 8 hr of informal conversations between the chief oncologist with 20 years of profes-sional experience and other physicians from oncology, hematology, anesthesiology, surgery, andnephrology. Conversations were audio recorded during the shadowing of the chief oncologist (seeCzarniawska-Joerges, 2007). The data were recorded in the courtyard of the hospital, where thephysicians met before the beginning of the workday, or in the hospital wards. More than halfof the conversations had to do with patients’ clinical cases. Other issues covered organizational,bureaucratic, or logistical matters or, to a minimal extent, the doctors’ social lives, and poli-tics. The focus of these routine conversations2 were “nonstandard” clinical cases that activatedcollaborative diagnostic practices.

Conversations were transcribed verbatim, following a simplified version of the Jefferson’s(2004) transcription system. The analysis moved through the following steps: (a) recurrent andindependent readings of transcripts informed by a broad understanding of the concept of prac-tice, as conceived in the theoretical introduction of the article; (b) development of interpretativecategories of diagnostic decision making that took into account the features of the specializedmedical domain of oncology. The development of interpretive categories followed what Scribner(1986) called a “functional” analysis conducted to grasp the “what-for of thinking, to examinehow thinking is related to doing” (p. 16); and (c) highlighting empirical evidence for categoriesin the transcripts.

ANALYSIS

We recurrently identified in the data set: (a) practices of joint interpretation, (b) practices ofintersubjective generation and validation of hypotheses, and (c) practices of postponing thediagnostic decision. As visible in the transcripts, such practices are intertwined in the ongoingdevelopment of the interaction; however, we distinguish them in the analyses to focus on theirspecific contribution to the diagnostic decision making process. In what follows, we present theanalyses of two episodes illustrating the complexity of thinking and doing in a hospital medicalcommunity.

Episode 1: “So Maybe It’s Just a Cancer”

In the first episode of diagnostic reasoning and decision making, “So Maybe It’s Just a Cancer”the chief oncologist and a medical internist talk about a patient (a woman) hospitalized in thedepartment of nephrology. They have been asked to provide a consultation. After their talk, the

1Among the hospital’s departments were psychiatry, intensive care, urology, surgery, nephrology, cardiology, generalmedicine, orthopaedics, and others.

2Interviews with doctors confirmed that such conversations are very common during the hospital’s everyday activities.

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

6 ALBY, ZUCCHERMAGLIO, AND BARUZZO

oncologist reports their conclusions to the nephrologist, as we see in the second excerpt. Thepatient shows spots in the lungs of uncertain clinical evaluation. The excerpt begins when theoncologist asks his colleague if he has been contacted by the nephrologist “about that patient”(see Excerpt 1):

Excerpt 1: “So Maybe It’s Just a Cancer”

Participants: ONC oncologist; MED medical internist

1. ONC: one more thing (0.2) Did Rossi ((the nephrologist)) happen to call you yesterday aboutthat patient?

2. MED: yes.

3. ONC: but is she neoplastic3 for you? (0.2) for sure?

4. MED: no.

5. ONC: exactly (0.2) indeed, I believe- (.) but-

6. MED: ()

7. ONC: m = hm. that’s what I thought (0.2) that those things on the skin could be a streptococcus4

8. MED: a streptococcus or a ()

9. ONC: m = hm. (0.2) that’s what I was thinking, indeed I told him that when I was-

10. because I told him a possibly neopla:stic patient (0.2) but not necessarily at all (.)

11. but I said (0.2) I said ∗it is stra:nge all these-∗ (0.2) ((stains, authors’ note))

12. if it was a cancer (0.2) all these metastases in the lungs ((would be impossible, authors’ note.))

13. tut ((shakes his head)) she has something like (0.5) it reminded me of (.) cryptoccoccus.5

14. MED: could be.

15. ONC: So I still know my medicine, right?

The social dimension of the achievement of a cancer diagnosis becomes visible when the dailyinteractions among physicians are analyzed. An interesting finding is that the diagnosis is notcarried out only by the oncologist but developed as a practice of joint interpretation that involvesdifferent doctors with different expertise. Specialists share concerns and try to solve them throughsophisticated forms of talk that support diagnostic decision making within an interdisciplinaryfield such as oncology.

What precisely is the contribution of the conversation between the oncologist and the internistwith respect to the diagnosis? We first note that the diagnosis is not a quick and easy process:There is no immediate recognition or classification of obvious symptoms, but rather the initiationof a process of clinical reasoning “out loud.” This practice enables the social distribution ofinterpretive resources while yielding a richer knowledge repertoire that is put to the service ofclinical reasoning.

3Neoplastic means related to a tumor or neoplasia.4Streptococcus is a group of bacteria that may cause infections.5Cryptococcus is a fungus.

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

DIAGNOSTIC DECISION MAKING IN ONCOLOGY 7

It should be noted that the internist is required to provide an assessment of the tumor(turn 3) and that it is the oncologist who exhibits expertise as a medical internist (turn 13).This shows that there is a partial overlap between the skills of the two specialists—probablydue to several factors, including their training and previous experiences—that is the premise andoutcome of their everyday practice of collaboration. At the same time, the epistemic author-ity that each of them possesses with regard to their respective fields of medical specializationis visible in the conversation from the way in which the internist is considered to have privi-leged access to general medical expertise, such as reference to the Cryptococcus, and therefore isable to validate the oncologist’s hypothesis (“So I still know my medicine, right?”; turn 15).Similarly, the oncologist gives an assessment of the neoplasia (turns 11–12), thus implyingthat he has a special “right” (Heritage & Raymond, 2005) to assess the neoplastic issue andexercising his ability to categorize events and to make specific knowledge on cancer mattersavailable.

Let us now examine how such a joint interpretation develops. The conversation between thetwo doctors actually starts from the conclusions they have reached (turns 3–5) and goes on toprovide the elements that supported the reasoning. They conduct a brief verification of theirdiagnostic hypotheses and then retrace together the information from which the conclusions havebeen reached. This process not only serves to justify the doubts on the neoplastic assumption, butalso allows the generation of a new diagnostic hypothesis (the cryptococcus).

The confutation of the neoplastic hypothesis (turns 3–4) leads to guesswork formulated bythe oncologist as personal opinions (turns 5, 7, 9). The argumentative force and plausibilityof these speculations are increased by being shared and by having undergone this moment ofintersubjective validation. It is noteworthy that in reasoning aloud, some items of information,rather than others, are selected as relevant for diagnostic purposes (“those things on the skin,”turn 7; spots in the lungs similar to pulmonary metastases). These operations of “highlighting”(Goodwin, 1994) and classification (Bowker & Star, 1999; Goodwin, 1994, 1999) identify thesigns and clues that may match with models of known illness or, in a comparative way, withcharacteristics of patients previously seen. This manner of arguing is similar to what in logic iscalled “abductive inference.” Abductive reasoning is an inference that yields the best availableexplanation (the philosopher, Charles Peirce, first called it “guessing”). One hypothesis is favoredover another because of its capacity to account for events otherwise unexplainable. Through thesediscursive inferential practices, doctors jointly try to build an interpretative relationship between,often contradictory, clinical evidence and known diseases (turn 7, turn 12).

In contrast with a model of linear rationality, we see that neither of the alternatives is discardedand that the conversation between oncologist and internist does not lead, for the moment, to thechoice of the best hypothesis. Although the internist indeed shares the oncologist’s doubts aboutthe diagnosis of cancer, the alternative explanatory hypothesis on the spots on the lungs—a diag-nosis of Cryptococcus—is still commented on by the internist with “could be.” The frequent useof mitigators and dubitative formulations (turns 7, 10, 14) shows that the diagnostic interpretationis still uncertain and that the discussion between the two doctors is not conclusive. Such doubtfulformulations are rather common in our data set and, as we have seen here, they support a practiceof hypothesis generation in cooperative diagnostic work (but see Lehtinen, 2013, for differentfunctions performed within genetic counseling).

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

8 ALBY, ZUCCHERMAGLIO, AND BARUZZO

After about 10 min the oncologist goes into the nephrology ward to report the outcome—still provisional but validated with the internist—of their diagnostic work on the patient. Thisconversation supplements the written advice that the oncologist has given to the nephrologist:

Excerpt 2

(10 min later, the oncologist goes to the nephrologist, NEF)

1. ONC: a gift? ((NEF gives him an envelope)) because (0.5)

2. also Cuomo ((medical internist)) has doubts about that- about the fact that that patient

3. could be <neopla:stic> (0.5) because I’ve looked at her (0.2)

4. I said, actually having written it in the report, I wrote

5. ((speaks while reading the envelops)) SO:: on the advice

6. I wrote that this patient is (0.2) po:ssibily neoplastic. (0.5) but.

7. the other idea I had was that it could be () just that.

8. NEF: m:: ((nods))

9. ONC: because she ((the patient, N.d.A.)) is stra:nge to be a cancer. She doesn’t even havea liver metastasis6 (0.5)

10. but she’s yellow.

11. NEF: m:: ((nods))

12. ONC: I don’t know (0.5) then she has some cu:rious skin lesions (.)

13. that remind me a lot of cryptococcus

14. NEF: m:. ((nods))

15. ONC: I don’t know (0.2) maybe then it’s not like this (0.5) ◦but here◦ I had this idea.

16. so maybe it’s just a cancer (0.2) but (.) in short, oh well

Compared with the written advice, the conversation between the oncologist and the nephrologistis much more complex. The oncologist cites the exchange with the internist in support of theplausibility of the neoplastic hypothesis (turn 2). The exchange that occurred with the internistexemplifies the practice of validation of hypothesis through which doctors can assess the robust-ness of their arguments and arrange subsequent actions. Thanks to such a practice, the oncologistcan now propose to the nephrologist, with enough confidence, an alternative diagnostic hypoth-esis (turn 7) and strongly to support the reasons that refute the cancer hypothesis: at turn 9, hesays, “because she’s strange to be a cancer.” This evaluative allegation is presented as a matterof fact, with independent certainty and validity. It reveals the premise on which the reasoning isbased, whose diagnostic conclusions have already been formulated in the previous turns (turns2–3, 6–7), at the very beginning of the encounter.

It should be also noted that in these two excerpts diagnostic talk shows, differently from otherstudies such as Antaki et al. (2005), an organization in which the “implication” comes first,

6Metastasis is the spread of a cancer from one organ or part to another nonadjacent organ or part.

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

DIAGNOSTIC DECISION MAKING IN ONCOLOGY 9

followed by the evaluations that led to the diagnostic outcome. This backward sequential orga-nization of the argumentation seems to support the local purposes of their conversation (i.e.,obtaining an independent judgment from the internist and avoiding “confirmation bias”; account-ing for the requested advice and report). The judgment against the initial hypothesis is furthersupported with accounts suggesting a causal, but contradictory, link with the clinical assessment(turns 9–10) and partly compatible with other explanatory frameworks (turns 12–13). Numerousdubitative expressions (turn 12, turn 15) mitigate the validity and plausibility of the differentdiagnostic hypotheses, maintaining uncertainty.

We note here how the expertise distributed in the professional community is a resource usedto account for the reliability of the available diagnostic hypotheses. Moreover, we see that theinteraction between oncologist and nephrologist ends without a clear position being taken, sothat all the diagnostic hypotheses are still feasible, including the initial one on cancer. This latteris recalled toward the end, and it is described as the simplest diagnosis (turn 16: “so maybeit’s just a cancer”), in that it makes it possible to ignore the relationship with the complex andlittle-known configurations of the clinical evidence.

A time may come when, outside our present frame, a decision will be taken on how to treatthe patient hospitalized in nephrology. Let us note, however, the temporal organization of thisprocess. The diagnosis observed here has a provisional status. There is therefore a temporal evo-lution of the diagnosis, and there is a stage, particularly central to diagnostic reasoning, in whichmultiple interpretations are brought into play. A great deal of medical and nonmedical informa-tion is collected and selectively highlighted, creating an interpretative framework of increasingcomplexity.

This organization, which is unforeseen in the sequence of reasoning development on a logicof abstract rationality, seems to support a practice of postponing the decision, gaining time inwhich to collect and collate information, and in which events can be clarified, thereby reducingthe possibility of making mistakes. Hence this is not an “empty” time, but one filled with, amongother things, tests and examinations, exchanges with other physicians, and joint argumentationsthat support the processing and validation of information, producing as a (intermediate) resultan increase in the complexity of the diagnostic picture. This collaborative practice appears tohelp to avoid diagnostic errors. In particular, it reduces, through the collaborative generation ofhypotheses, the frequent mistake (Cartabellotta, 2001) of interpreting data as confirmation of theinitial hypothesis.7 Our analysis provides another illustration and validation of how systems ofdistributed cognition and communities of practice organize themselves in order to reduce risks oferror in complex situations (see Hutchins, 1993).

Episode 2: “A Heavy Smoker”

The importance of talk among physicians in diagnostic work is highly visible in Excerpts3, 4, and 5, in which the oncologist and a surgeon discuss possible diagnoses for a patient.

7Studies on large samples of cases in various branches of medicine have found an incidence of serious diagnosticerrors of about 15% to 20%, half of which have a probable impact on prognosis (Podbregar et al., 2001; Shojania,McDonald, Wachter, & Owens, 2004). In the United States, one fifth of lawsuits on medical issues concern diagnosticerrors (Bartlett, 1998).

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

10 ALBY, ZUCCHERMAGLIO, AND BARUZZO

Biographical details of the patient, assessment of his health, examination results, theoreticalstatements regarding medical knowledge are intertwined in a narrative that leads to the productionof several hypotheses and to the decision to proceed with a physical examination:

Excerpt 3: “A Heavy Smoker”

Participants: ONC oncologist; SUR surgeon

1. ONC: on the issue that concerns you (0.2) what did you decide?

2. SUR: which of the many? (.) because there are a lot

3. ONC: of the relative

4. SUR: what do I know (0.2) now we’re wai[ting

5. ONC: [mah <ade:nocarcino:ma8> (0.5)

6. if it really is that (0.5) if he is someone who is (.) fine = fine = fine (0.5) you can also::

7. SUR: think that he even came fishing with me

8. ONC: you can also spare him ((avoiding treatments, authors’ note.))

9. SUR: but it is very diffi[cult

10. ONC: [an 84-year-old who is in good health (0.5) I mean.

The oncologist asks the surgeon about a patient who has some degree of kinship with thesurgeon, hence the initial misunderstanding on an issue that concerns him (turns 1–3). Althoughthe question asks for a decision, the response expresses doubt (turn 4). The oncologist formulatesthe current diagnosis and then immediately reduces its plausibility (turn 6), revealing it as a guessand making it provisional. Such an uncertain formulation leads to the practice of intersubjectivegeneration and validation of hypothesis that allows doctors to build a scenario through whichthey can imagine possible courses of action (turn 8).

The words of the oncologist (turn 6) are received by the surgeon as a request for evaluation ofthe patient’s overall health, to which he responds by introducing items of nonmedical information(turn 7) that make it possible to put the patient in the category of “84-year-olds who are in goodhealth.” What is interesting about this categorization of the patient’s social identity is its impacton subsequent decisions concerning treatment (turn 8), further underlining the function of thesenarratives as places for the design of medical activities.

Another effect of the formulation of the diagnosis of adenocarcinoma is to “invite” hypothesisgeneration (i.e., thyroid cancer), as we see in the continuation of the physicians’ conversation:

Excerpt 4

11. SUR: you know right (0.2) because (.) I was doing tests to see if it was thyroid

12. ONC: hm hm.

13. SUR: like this (0.2) just to [test

14. ONC: [yes (.) you told me

8Adenocarcinoma is a cancer of epithelial tissue.

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

DIAGNOSTIC DECISION MAKING IN ONCOLOGY 11

15. SUR: hm hm. and maybe today

16. ONC: however in my life (.) I’ve never se:en (0.5) a cancer stra- that makes- that is (0.2)

17. the thyroid is the one thing that you always suspect and it never is

18. SUR: no I’ve seen them

19. ONC: yes, no, me too I’ve seen thyroid cancers

20. but every time I started with the th[yroid cancer

21. SUR: [ah yes

22. ONC: all the times you thought that it was something else, that you said

23. ∗well let’s exclude that it is thyroid∗ (0.2) it has never been.

24. SUR: ◦ah yes◦(0.2) okay=okay

The surgeon does not provide the oncologist with any brand-new information (see turn 14).Therefore this communication goes beyond the sole purpose of informing: It creates the opportu-nity for practices of joint interpretation in which different but overlapping knowledges can worktogether.

The surgeon’s communication elicits a comment by the oncologist on the diagnosis of thyroidin general. The oncologist’s premise epistemically grounds the story by using his long, profes-sional experience as a “warranty” (turn 16). He then increases the status of validity and generalityof his opinion by presenting it as a certainty, as a “normal” rule to which to refer in professionalpractice (turn 17). The surgeon rejects the assertion and cites his experience as evidence (turn 18).This leads to a more detailed, limited, and contextualized reformulation by the oncologist (turns19–24), with which the surgeon agrees. The oncologist, who now again speaks in first person,makes it clear that (a) the “real” thyroid cancers are those where thyroid cancer was the startinghypothesis, and (b) when thyroid cancer was a residual hypothesis, then the diagnostic processhas always confirmed that it was something else.

The normative character of diagnostic practice is built in contested interactions like these, inwhich, as we see, the agreement is not granted, but requires sharing and negotiating a practicalknowledge and the beliefs that shape it. Thanks to the activation of a diagnostic community andthrough “war stories” like this one (Orr, 1996), practical knowledge can be formulated, shared,and checked, becoming an interpretative resource for future diagnostic interactions. Besidesthyroid cancer, later in the conversation, the two doctors formulate other diagnostic hypotheses:

Excerpt 5

25. ONC: oh well anyhow in [short

26. SUR: [so you say that is not- that it is useless to do something?

27. ONC: no=no (.) one moment. (0.5) it’s not a prostate, right?

28. SUR: no.(0.2) I don’t think so.

29. ONC: did you do his PSA9 ?

9PSA is a screening test for prostate cancer.

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

12 ALBY, ZUCCHERMAGLIO, AND BARUZZO

30. ((SUR nods))

31. ONC: that’s why I said ∗let me examine him∗ (0.2) I’m not gonna kill him

32. if I examine him by hand? (0.2) I’ll say that I am the ear specialist, the pediatrician,

33. SUR: yes

34. ONC: the gynecologist,

35. SUR: yes, right

36. ONC: I can even see him (0.5) on the day that I’m on call (0.5) so in a setting,

37. in a non=non-cancer setting

38. SUR: because we have to to speak to Como ((medical internist)) I think he did it to him

39. ONC: of course he did it to him. (0.5) then obviously it won’t be.

40. SUR: I thought of the prostate as well. but I noticed () pulmonary, the lymph nodes,

41. ONC: ◦it will be a lung cancer◦ (0.2) does he smoke then?

42. SUR: yes.

43. ONC: then that’s it.

44. SUR: a heavy smoker.

Faced with the possibility of doing nothing (diagnostic failure), the oncologist restarts theprocess of hypotheses generation (i.e., prostate and lung cancers). Prostate cancer is considerednot very likely (turn 28), but it is still checked and kept available (though with caution; turn 39).There follows a humorous dialogue in which a physical examination is proposed (turns 31–37).The surgeon’s concerns regarding the patient’s psychological protection are made explicit and,through irony, signalled as exaggerated (turn 31), but taken into account (turns 36–37). A clinicalobservation by the surgeon about the lungs and lymph nodes allows the oncologist to formulatea fourth hypothesis (a fifth, if one includes that of doing nothing), which is lung cancer. Thishypothesis greatly increases in likelihood after the information that the patient smokes, to thepoint of being formulated as a certainty by the oncologist (turn 43), and further reinforced by thesurgeon (turn 44).

The argumentative strength with which the last hypothesis is presented might induce us tosuppose that this is the definitive diagnosis for this patient. However, it should be borne in mindthat the function of these practices of joint interpretation is not to reveal the correct diagnosis,in spite of their definitive formulation. Also here in the “smoker’s case,” other conversations anddiagnostic practices, including the physical examination, and still awaited test results relativeto the other hypotheses, are necessary before arriving at the final decision. As in other profes-sions, these conversations enable the doctors to imagine and explore possible courses of action inorder eventually to reach a decision (Alby & Zucchermaglio, 2007, 2008; Newmann, 1998; Ochs,Gonzales, & Jacoby, 1996). Such practices of joint interpretation give the diagnostic hypothesis,not an abstract possibility but a concrete and situated one: They are in fact modeled on the indi-vidual patient, within “dense” interpretative frameworks that combine medical and nonmedicalknowledge, like fishing or smoking. The patient appears at the beginning as “an 84-year-old whois fine,” is later represented through the organs of his body, those possibly affected by the disease

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

DIAGNOSTIC DECISION MAKING IN ONCOLOGY 13

(“a possible prostate,” lung, lymph nodes), and then returns as a “whole” described as “a heavysmoker” in the final part of the interaction.

This bricolage helps to keep different agencies and information together in the same interpre-tative framework, combining the results of a thorough check on parts of the patient’s physicalbody with the autonomous and independent behaviours of the patient as a social body, includ-ing his habits and his lifestyle. Such practices increase the interpretative resources and createcomplex scenarios in which to reason while postponing the final diagnostic decision.

CONCLUSIONS

There are few studies on diagnostic decision making that focus directly on oncology. Our arti-cle has contributed to the gap in the literature by providing a unique empirical analysis of howdiagnostic decision making is accomplished in naturally occurring medical interactions. Our anal-ysis shows that doctors rely on three collaborative practices to organize the diagnostic decisionmaking process. Let us discuss them in further detail.

Practices of Joint Interpretation

Our analysis has shown that an essential part of diagnostic reasoning happens “aloud” and isdistributed in the interactions among the actors of a diagnostic community. As we have seen inthe episodes analyzed, the activation of a diagnostic community provides access to diverse typesof expertise, information, and clinical data about the patient while it allows the maintenance anddevelopment of repertoires of practical knowledge to which the doctors refer in a shared manner.We have seen how doctors repeat information and hypotheses that have already been mentionedpreviously, thus showing the importance of practices of joint interpretation for the ongoing diag-nostic work. What doctors bring to these practices is a long experience of similar situations thatserves as their toolbox for formulating hypotheses and making decisions. An interesting pointhere is how talk and reasoning are closely intertwined, mutually dependent in the developmentof interpretative practices. Through such practices doctors accomplish, at a collective level, thematching of the clinical evaluation with the known models of the illness.

Practices of Intersubjective Generation and Validation of Hypothesis

Doctors build argumentative contexts in which hypotheses can be generated and tested, assess-ing the robustness of their arguments and arranging subsequent actions (e.g., further exams andconsultations). These argumentative practices are the “creative engine” whereby the hypothesesincrease—like the five hypotheses in the second episode—and interweave with other ongoingmedical practices. It is thanks to the embeddedness within a texture of hospital activities thatthe formats of doctors’ joint hypothetical reasoning do not follow the sequential pattern of linearrationality but exhibit a reverse sequence in which the diagnostic outcome is the first thing to becommunicated (see Episode 1).

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

14 ALBY, ZUCCHERMAGLIO, AND BARUZZO

Another important difference with respect to models of linear rationality is that hypothesesare, for the time being, not really “alternatives.” There is no best hypothesis that is immediatelychosen over the others. They are all at stake, although with different degrees of probability.

Practices of Postponing the Diagnostic Decision

Doctors do not rush into a definitive decision. Rather, they gain time in which to collect andcollate information, to make a number of diagnostic hypotheses with provisional status, and toengage in joint argumentations that validate and complete information, thereby reducing the pos-sibility of making errors. In the data that we analyzed, the diagnosis evolved over time and wentthrough a phase of impasse of variable duration across the data set. In such a phase of impasse,a great deal of clinical information is highlighted, collected, and put into play; hypotheses areformulated and kept available; and an increasingly complex diagnostic picture is created. This isdifferent from what is recommended by rationalist models that emphasize reduction of informa-tional entropy, and cognitive strategies for certainty, but similar to recent observations on humancognition proceeding through cycles of disorder and reorder (Hutchins, 2012).

Postponing the diagnostic decision is thus a practice that helps in dealing with the complex-ity and uncertainty that characterizes oncology as a branch of medicine. In this scenario, whatat a first glance may be seen as a doctor’s indecisiveness reflects the degree of “relative knowl-edge” of current medical science and, therefore, the doctor’s experience and familiarity with thatknowledge made of situated certainties (Fox, 1957; Montgomery, 2006).

Our analyses describe the kind of situated rationality that inform decision making whenobserved in real-world settings. Compared to the three key defining notions of decision mak-ing in the model of linear rationality described in the introduction, the practices analyzed displaydifferent features. First, diagnostic decision making shows a temporal evolution, which includesa stage in which uncertainty is at its maximum, hypotheses proliferate, and informational com-plexity increases. Second, decision making does not follow a sequential pattern of rational andlinear causality (“if–then”) but relies on a practical logic that account for the interlink withother work practices (such as accounting for a consultation) and accordingly shapes the orga-nization of the argumentation (which in this case became “then–if”). Third, diagnostic decisionmaking results are shaped by the local constraints of the specialized medical domain and theopportunities of the context. In particular we observed how, in this interdisciplinary hospital,doctors exploit the distributed knowledge of their diagnostic community to make hypotheticalreasoning easier to master in a complex domain such as oncology, overcoming limits in indi-vidual knowledge and cognition. So doing, doctors follow an economic logic that optimizesefforts. The solutions observed are sorts of “translations into practice” of the inferential prin-ciples and theories of medicine as a formalized system of knowledge in ways that take accountof local resources and constraints, limits to the doctors’ knowledge and mental abilities, and thesocial organization of work within professional communities.

More generally, our article, drawing upon the concept of practice, shows an approach tounderstanding and improving decisions-in-context. A better understanding of the practices thatorganize decision making can be used to support existing medical diagnostic communities orinform the practice-driven redesign of work processes while, more in general, improving ourknowledge of the social organization of cognition in complex domains of activity.

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

DIAGNOSTIC DECISION MAKING IN ONCOLOGY 15

ORCID

Francesca Alby http://orcid.org/0000-0001-5126-5293

REFERENCES

Alby, F., & Zucchermaglio, C. (2006). “Afterwards we can understand what went wrong, but now let’sfix it.” How situated work practices shape group decision making. Organization Studies, 27, 943–966.doi:10.1177/0170840606065703

Alby, F., & Zucchermaglio, C. (2007). Embodiment at the interface: Materialization practices in web design. Researchon Language and Social Interaction, 40, 255–277. doi:10.1080/08351810701354714

Alby, F., & Zucchermaglio, C. (2008). Collaboration in web design: Sharing knowledge, pursuing usability. Journal ofPragmatics, 40, 494–506. doi:10.1016/j.pragma.2007.10.008

Alby, F., & Zucchermaglio, C. (2009). Time, narratives and participation frameworks in software troubleshooting.Computer Supported Cooperative Work (CSCW), 18, 129–146. doi:10.1007/s10606-008-9090-7

Antaki, C., Barnes, R., & Leudar, I. (2005). Diagnostic formulations in psychotherapy. Discourse Studies, 7, 627–647.doi:10.1177/1461445605055420

Bartlett, E. (1998). Physicians’ cognitive errors and their liability consequences. Journal of Healthcare Risk Management,18(4), 62–69. doi:10.1002/jhrm.5600180407

Beach, W. A. (Ed.). (2012). Handbook of patient–provider interactions: Raising and responding to concerns about life,illness, and disease. Cresskill, NJ: Hampton Press.

Boden, D. (1994). The business of talk: Organizations in action. Cambridge, England: Polity Press.Bowker, G. C., & Star, S. L. (1999). Sorting things out: Classification and its consequences. Cambridge, MA: MIT Press.Brookes-Howell, L. C. (2006). Living without labels: The interactional management of diagnostic uncertainty in the

genetic counselling clinic. Social Science & Medicine, 63, 3080–3091. doi:10.1016/j.socscimed.2006.08.008Button, G., & Sharrock, W. W. (1998). The organizational accountability of technological work. Social Studies of Science,

28, 73–102. doi:10.1177/030631298028001003Cartabellotta, A. (2001). Ragionamento diagnostico, Evidence-based Medicine e linee guida [Diagnostic reasoning.

Evidence-based medicine and guidelines]. Rivision of Medical Laborator – -JLM, 2, 1.Chaiklin, S., & Lave, J. (Eds.). (1993). Understanding practice: Perspectives on activity and context. Cambridge,

England: Cambridge University Press.Cicourel, A. V. (1985). Raisonnement et diagnostic: le rôle du discours et de la compréhension clinique en médecine.

Actes de La Recherche En Sciences Sociales, 60, 79–89. doi:10.3406/arss.1985.2292Cicourel, A. V. (2002). La gestion des rendez-vous dans un service médical spécialisé [The management of encounters

in a specialized medical service]. Actes de La Recherche En Sciences Sociales, 143, 3–17.Cole, M. (1996). Cultural psychology: A once and future discipline. Cambridge, MA: Harvard University Press.Czarniawska-Joerges, B. (2007). Shadowing and other techniques for doing fieldwork in modern societies. Frederiksberg,

Denmark: Copenhagen Business School Press.Engeström, Y., Miettinen, R., & Punamäki, R.-L. (1999). Perspectives on activity theory. Cambridge, England:

Cambridge University Press.Engeström, Y., & Sannino, A. (2012). Concept formation in the wild. Mind, Culture, and Activity, 19, 201–206.

doi:10.1080/10749039.2012.690813Falmagne, R. J. (2000). Positionality and thought: On the gendered foundations of thought, culture and development. In

P. H. Miller & E. Scholnick (Eds.), Developmental psychology through the lenses of feminist theories (pp. 191–213).New York, NY: Routledge.

Fiske, S., & Taylor, S. (1991). Social cognition (2nd ed.). New York, NY: McGraw-Hill.Fox, R. C. (1957). Training for uncertainty. In R. K. Merton, G. Reader, & P. L. Kendall (Eds.), The student-physician:

Introductory studies in the sociology of medical education (pp. 207–241). Cambridge, MA: Harvard University Press.Gherardi, S. (2012). Docta ignorantia: Professional knowing at the core and at the margins of a practice. Journal of

Education and Work, 25, 15–38. doi:10.1080/13639080.2012.644902Goodwin, C. (1994). Professional vision. American Anthropologist, 96, 606–633. doi:10.1525/aa.1994.96.3.02a00100

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

16 ALBY, ZUCCHERMAGLIO, AND BARUZZO

Goodwin, C. (1999). Practices of colour classification in professional discourse. In A. Jaworski & N. Coupland (Eds.),The discourse reader (pp. 474–491). London, England: Routledge.

Goodwin, C., & Goodwin, M. H. (1996). Seeing as a situated activity: Formulating planes. In D. Middleton & Y.Engeström (Eds.), Cognition and communication at work (pp. 61–95). Cambridge, England: Cambridge UniversityPress.

Han, P. K. J., Klein, W. M. P., & Arora, N. K. (2011). Varieties of uncertainty in health care: A conceptual taxon-omy. Medical Decision Making: An International Journal of the Society for Medical Decision Making, 31, 828–838.doi:10.1177/0272989X10393976

Heath, C. (1992). The delivery and reception of diagnosis in the general practice consultation. In P. Drew & J. Heritage(Eds.), Talk at work: Interaction in institutional settings (pp. 235–267). Cambridge, England: Cambridge UniversityPress.

Heritage, J., & Maynard, D. W. (Eds.). (2006). Communication in medical care: Interaction between primary carephysicians and patients. Cambridge, England: Cambridge University Press.

Heritage, J., & Raymond, G. (2005). The terms of agreement: Indexing epistemic authority and subordination in talk-in-interaction. Social Psychology Quarterly, 68, 15–38. doi:10.1177/019027250506800103

Hutchins, E. (1993). Learning to navigate. In S. Chaiklin & J. Lave (Eds.), Understanding practice: Perspectives onactivity and context (pp. 35–63). Cambridge, England: Cambridge University Press.

Hutchins, E. (1995). Cognition in the wild. Cambridge, MA: MIT Press.Hutchins, E. (2012). Concepts in practice as sources of order. Mind, Culture, and Activity, 19, 314–323.

doi:10.1080/10749039.2012.694006Hutchins, E., & Klausen, T. (1996). Distributed cognition in an airline cockpit. In Y. Engeström & D. Middleton (Eds.),

Cognition and communication at work (pp. 15–34). Cambridge, England: Cambridge University Press.Jefferson, G. (2004). Glossary of transcript symbols with an introduction. In G. H. Lerner (Ed.), Conversation analysis:

Studies from the first generation (pp. 13–23). Amsterdam, the Netherlands: John Benjamins.Koenig, C. J. (2011). Patient resistance as agency in treatment decisions. Social Science and Medicine (1982), 72,

1105–1114. doi:10.1016/j.socscimed.2011.02.010Kostopoulou, O., Russo, J. E., Keenan, G., Delaney, B. C., & Douiri, A. (2012). Information distortion in physicians’

diagnostic judgments. Medical Decision Making, 32, 831–839. doi:10.1177/0272989X12447241Lave, J. (1988). Cognition in practice: Mind, mathematics, and culture in everyday life. Cambridge, England: Cambridge

University Press.Lave, J. (1993). The practice of learning. In S. Chaiklin & J. Lave (Eds.), Understanding practice: Perspectives on activity

and context (pp. 35–63). Cambridge, England: Cambridge University Press.Lave, J., & Wenger, E. C. (1991). Situated learning: Legitimate peripheral participation. Cambridge, England:

Cambridge University Press.Lehtinen, E. (2013). Hedging, knowledge and interaction: Doctors’ and clients’ talk about medical information and client

experiences in genetic counseling. Patient Education and Counseling, 92, 31–37. doi:10.1016/j.pec.2013.02.005Lipshitz, R., Klein, G., & Carroll, J. S. (2006). Introduction to the special issue. Naturalistic decision mak-

ing and organizational decision making: Exploring the intersections. Organization Studies, 27, 917–923.doi:10.1177/0170840606065711

March, J. G. (1991). Exploration and exploitation in organizational learning. Organization Science, 2, 71–87.doi:10.1287/orsc.2.1.71

March, J. G. (1994). A primer on decision making: How decisions happen. New York, NY: Free Press.Maynard, D. W. (1992). On clinicians co-implicating recipients’ perspective in the delivery of diagnostic news. In P. Drew

& J. Heritage (Eds.), Talk at work: Social interaction in institutional settings (pp. 331–358). Cambridge, England:Cambridge University Press.

Miettinen, R., Paavola, S., & Pohjola, P. (2012). From habituality to change: Contribution of activity the-ory and pragmatism to practice theories. Journal for the Theory of Social Behaviour, 42, 345–360.doi:10.1111/j.1468-5914.2012.00495.x

Montgomery, K. (2006). How doctors think: Clinical judgment and the practice of medicine. New York, NY: OxfordUniversity Press.

Motterlini, M., & Crupi, V. (2005). Decisioni mediche. Un punto di vista cognitivo [Medical decisions. A cognitive pointof view]. Milan, Italy: Raffaello Cortina Editore.

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

DIAGNOSTIC DECISION MAKING IN ONCOLOGY 17

Newmann, W. W. (1998). Foreign policy decision making, garbage cans, and policy shifts: The Eisenhower admin-istration and the “Chances for Peace” speech. The American Review of Public Administration, 28, 187–212.doi:10.1177/027507409802800205

Nicolini, D. (2012). Practice theory, work, and organization: An introduction. Oxford, England: Oxford UniversityPress.

Ochs, E., Gonzales, P., & Jacoby, S. (1996). “When I come down I’m in the domain state”: Grammar and graphic repre-sentation in the interpretive activity of physicists. In E. Ochs, E. A. Schegloff, & S. A. Thompson (Eds.), Interactionand grammar (pp. 328–369). Cambridge, England: Cambridge University Press.

Oliver, D., & Roos, J. (2005). Decision making in high velocity environments: The importance of guiding principles.Organization Studies, 26, 889–913. doi:10.1177/0170840605054609

Opel, D. J., Heritage, J., Taylor, J. A., Mangione-Smith, R., Salas, H. S., DeVere, V., . . . Robinson, J. D. (2013).The architecture of provider–parent vaccine discussions at health supervision visits. Pediatrics, 132, 1037–1046.doi:10.1542/peds.2013-2037

Orr, J. E. (1996). Talking about machines: An ethnography of a modern job. Ithaca, NY: Cornell University Press.Peräkylä, A. (1998). Authority and accountability: The delivery of diagnosis in primary health care. Social Psychology

Quarterly, 61, 301–320. doi:10.2307/2787032Pilnick, A., Hindmarsh, J., & Gill, V. T. (2009). Beyond “doctor and patient”: Developments in the study of healthcare

interactions. Sociology of Health & Illness, 31, 787–802. doi:10.1111/j.1467-9566.2009.01194.xPodbregar, M., Voga, G., Krivec, B., Skale, R., Pareznik, R., & Gabrscek, L. (2001). Should we confirm our clinical

diagnostic certainty by autopsies? Intensive Care Medicine, 27, 1750–1755. doi:10.1007/s00134-001-1129-xResnick, L. B., Säljö, R., Pontecorvo, C., & Burge, B. (Eds.). (1997). Discourse, tools and reasoning: Essays on situated

cognition. Berlin, Germany: Springer.Roth, E. M., Multer, J., & Raslear, T. (2006). Shared situation awareness as a contributor to high reliability performance

in railroad operations. Organization Studies, 27, 967–987. doi:10.1177/0170840606065705Schatzki, T. R. (2001a). Introduction: Practice theory. In T. R. Schatzki, K. D. K. Cetina, & E. V. Savigny (Eds.), The

practice turn in contemporary theory (pp. 10–23). London, England: Routledge.Schatzki, T. R. (2001b). Practice mind-ed orders. In T. R. Schatzki, K. D. K. Cetina, & Savigny, E. V. (Eds.), The practice

turn in contemporary theory (pp. 50–64). London, England: Routledge.Scribner, S. (1984). Studying working intelligence. In B. Rogoff & J. Lave (Eds.), Everyday cognition: Its development

in social context (pp. 9–40). Cambridge, MA: Harvard University Press.Scribner, S. (1986). Thinking in action: Some characteristics of practical thought. In R. J. Sternberg & R. K. Wagner

(Eds.), Practical intelligence: Nature and origins of competence in the everyday world (pp. 13–30). Cambridge, UK:Cambridge University Press.

Shojania, K. G., McDonald, K. M., Wachter, R. M., & Owens, D. K. (Eds.). (2004). Closing the quality gap: A criticalanalysis of quality improvement strategies (Vol. 1: Series overview and methodology). Rockville, MD: Agency forHealthcare Research and Quality.

Simon, H. A. (1957). Administrative behavior. New York, NY: Macmillan.Stivers, T. (2002). Participating in decisions about treatment: Overt parent pressure for antibiotic medication in pediatric

encounters. Social Science & Medicine, 54, 1111–1130. doi:10.1016/S0277-9536(01)00085-5Streeck, J., Goodwin, C., & LeBaron, C. (Eds.). (2011). Embodied interaction: Language and body in the material world.

Cambridge, England: Cambridge University Press.Szambok, C., & Klein, G. (Eds.). (1995). Naturalistic decision making. Hillsdale, NJ: Erlbaum.Toulmin, S. E. (1996). Concluding methodological reflections: Elitism and democracy among the sciences. In B.

Gustavsen & S. E. Toulmin (Eds.), Beyond theory: Changing organizations through participation (pp. 203–225).Amsterdam, the Netherlands: John Benjamins.

Vygotsky, L. (1990). Pensiero e linguaggio [Thought and language] (L. Mecacci, Ed.). Bari, Italy: Laterza.Wenger, E. C. (2002). Communities of practice. In N. J. Smelser & P. B. Baltes (Eds.), International encyclopedia of the

social & behavioral sciences (pp. 2339–2342). Oxford, England: Pergamon Press.Zucchermaglio, C., & Alby, F. (2012). Planning and assessing performance through narratives in soccer team meetings.

Qualitative Research in Sport and Exercise and Health, 4, 459–469. doi:10.1080/2159676X.2012.693523

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

18 ALBY, ZUCCHERMAGLIO, AND BARUZZO

APPENDIXExcerpts in Italian

Excerpt 1: “So Maybe It’s Just a Cancer”1 ONC: un’altra cosa (0.2) che per caso: t’ha chiamato ieri Rossi ((il nefrologo)) per quella ma:la:ta?2 MED: sì.3 ONC: ma è neoplastica secondo te ?(0.2) sicuramente?4 MED: no.5 ONC: Appu:nto (0.2) infatti. io=credo (.) ma-6 MED: ()7 ONC: è. quello pensavo io (0.2) che quelle cose alla pelle potrebbero essere uno streptococco,8 MED: uno streptococco o un ()9 ONC: è. (0.2) quello stavo a pensà io infatti gli ho detto che quando m’hanno-10 perché infatti gli=avevo=detto una paziente possibilmente neopla:stica (0.2) però non è detto affatto (.)11 però ho detto (0.2) ho detto ∗è stra:no tutte ste- ∗(0.2) ((macchie N.d.A.))12 se fosse un cancro (0.2) tutte ste metastasi polmonari ((sarebbero impossibili, N.d.A.))13 ntz ((cenno di diniego)) questa c’ha qualcosa de:: (0.5) a me ricordava (.) il criptococco14 MED: può darsi.15 ONC: Allora me la ricordo ancora la medicina, eh?

Excerpt 21 ONC: un regalo? ((NEF gli consegna una busta indirizzata a lui)) no perché (0.5)2 anche Cuomo ((internista)) esprime perplessità (.) su quella. sul fatto che quella paziente3 possa essere <neopla:stica> (0.5) perché io me la sono guardata (0.2)4 ho detto, essendo scritto sulla consule:nza, io ho scri:tto5 ((parla mentre legge ciò che è scritto sulla busta)) per=cui ALLORA sulla6 consulenza ho scritto che questo paziente è (0.5) possibilmente neoplastico. (0.5) ma7 l’altra=idea che mi ero fatto è che fossero () e ba:sta.8 NEF: m:: ((annuisce))9 ONC: perché è stra:na per essere un cancro. non c’ha manco una metastasi epatica (0.5)10 però è gialla.11 NEF: m:: ((annuisce))12 ONC: non lo so (0.5) poi c’ha delle curio:se lesioni cuta:nee (.)13 che a me mi ricordano tanto il criptococco14 NEF: m:. ((annuisce))15 ONC: non lo so. (0.2) magari non è così (0.5) ◦però ecco◦ m’ero fatto st’idea.16 magari poi è solamente un cancro (0.2) però (.) insomma vabbè

Excerpts 3–5: “A Heavy Smoker”1 ONC: per la questione che ti riguarda (0.2) poi che hai deciso?2 SUR: quale delle ta::nte? (.) perché sò parecchie3 ONC: del del parente4 SUR: e che ne so (0.2) mò sta stiamo aspetta[:ndo5 ONC: [mah <ade:nocarcino:ma> (0.5)6 se è veramente quello (0.5) se è uno che sta (.) bene=bene=bene (0.5) lo puoi pure::7 SUR: figurati che è venuto pure a pesca con me8 ONC: lo puoi pure graziare9 SUR: è ma è diffici[le10 ONC: [un ottantaquattrenne che sta be:ne (0.5) voglio=dì.11 SUR: sai no (0.2) perché (.) stavo facendo pro:ve che fosse tiroide12 ONC: è.13 SUR: così (0.2) giusto per pr[ova14 ONC: [sì (.) me l’hai detto

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015

DIAGNOSTIC DECISION MAKING IN ONCOLOGY 19

15 SUR: è. e forse oggi:::16 ONC: comunque in vita mia (.) non ho mai vi:sto (1.0) un cancro stra- che fa- cioè (0.2)17 la tiroide è una co:sa (.) che si sospetta sempre e non è mai.18 SUR: no io l’ho visti19 ONC: sì:: no, li ho visti anch’io i cancri alla tiroide (.)20 però:: tutte le volte che partivo con il cancro alla ti[ro- cioè21 SUR: [ah sì22 ONC: tutte le volte che pensavi che era un’altra cosa, che hai detto ∗

23 bè escludiamo che sia tiroide∗ (0.2) non era mai.24 SUR: ◦ah sì ◦(0.2) va bè=va bè25 ONC: va bè comunque non-[ insomma26 SUR: [quindi tu dici che no:::n (.) è inutile fare qualcosa?27 ONC: no=no (.)un momento. (0.5) non è una prostata, vero?28 SUR: e no.(0.2) non credo.29 ONC: pi esse a gliel’hai fa-?30 ((SUR annuisce con il capo))31 ONC: e per questo dicevo (.) ∗fammelo visità∗ (0.2) mica che te l’a- te=te l’accoppo (.)32 se lo guardo con le mani (0.2) io dico che faccio l’otori::no, il pedia::tra,33 SUR: sì34 ONC: il gineco::logo,35 SUR: sì va bè.36 ONC: lo posso pure guardare::(0.5) il giorno che sono di guardia. (0.5) cioè in ambiente,37 in ambiente non=non oncologico38 SUR: perché bisogna sentire Como ((medico internista)) credo che gliel’abbia fatto39 ONC: ◦sicuramente gliel’ha fatto (0.5) non sarà poi ovviamente.40 SUR: c’avevo pensato pure io alla prostata. però ho notato () polmonare, i linfonodi,41 ONC: ◦sarà un cancro al polmone◦ (0.2) lui fuma poi?42 SUR: è.43 ONC: e allora è que:llo.44 SUR: un grandissimo fumatore.

Dow

nloa

ded

by [

Fran

cesc

a A

lby]

at 1

1:05

23

Janu

ary

2015